1 82
��INTERNATIONAL JOURNAL OF AMERICAN LINGUISTICS
��VOL. I
��structural differences that exist between the separate groups. Thus, some very distinctive traits have been noted in the northern group, where a great number of nouns which, in the other groups, appear as independent words, are treated as suffixed stems. For that reason, a conclusive discussion of the probable genetic relationship between Chinook and Kalapuya will depend largely upon whether the distinct- ive traits of the northern dialects are due to bodily borrowing from the immediately ad- joining Chinook or whether they represent a survival of structural features that have be- come lost in the other dialects. Until this problem has been settled, we must refrain from jumping at any too hasty conclusions concerning the probable Chinook-Kalapuyan affiliations, no matter how tempting such con- clusions may appear.
Turning now to the Takelma-Kalapuyan aspect, the possibilities of a probable ultimate relationship are much stronger, although in this case also extreme caution must be exer- cised, at least for the time being. The struc- tural differences between these two languages are too great to be entirely wiped away be- cause of lexical correspondences of even the closest type. I shall mention only some of the most salient distinctive traits. Nominal in- corporation is lacking in Kalapuya, and pro- nominal incorporation is confined to the ob- ject. All subjective relations are expressed by means of the independent pronouns which precede the verb in the form of very loose pre- fixes; similarly possessive relations are ex-
��pressed by means of loose prefixes. No dis- tinction is made, in Kalapuya, by means of phonetic changes between aorist and non- aorist stems, tense being indicated by means of particles. On the other hand, there are evidences of structural elements which, when considered from a numerical point of view, would seem peculiarly typical of these two languages. Among these correspondences may be mentioned : the lack, in both languages, of a dual number; of an inclusive and exclusive person; the absence of nominal cases; the considerable use of end-reduplication and the total absence of initial reduplication; the presence of instrumental affixes denoting body-part nouns (in Kalapuya these appear as suffixed particles ; in Takelma as prefixes) ; and the apparent absence of distinct verb- stems for the singular and plural.
To sum up, while the correspondences that have been noted between Kalapuya and Chi- nook on one hand, and between Kalapuya and Takelma on the other hand, are too numerous and too close to be explained away by a theory of accident or recent borrowing, they are not conclusive enough to constitute adequate proof for a genetic relationship between these three linguistic stocks. Such an assumption, to be correct, must be predicated upon the introduction of additional material, especially from the field of Kalapuya linguistics. And I make bold to predict that additional data will be produced in the near future, for, it must be remembered, this is after all only a preliminary paper.
��BUREAU OF AMERICAN ETHNOLOGY WASHINGTON, D. C.
�� �
